Cerebral vascular malformations (CVMs) are prevalent in up to 4% of the population and are a common cause of stroke and epilepsy. Mechanisms predisposing to their genesis and progression have not been elucidated and their clinical course remains unpredictable. The candidate is a clinician- scholar in his 12th year post-residency training with focused interests in neurovascular surgery, having contributed seminal clinical-pathologic correlations and novel classification of CVMs. He has mentored students and residents in their first exposure to clinical research, resulting in numerous peer-reviewed publications by these trainees, several of whom are pursuing careers in academic neurosurgery. The proposed mid-career investigator award aims to expand and protect the candidate's committed time to patient-oriented research and mentoring of beginning clinician- investigators, within the framework of ongoing study on pathobiology of CVMs (R01NS36194) and other established collaborations. The candidate proposes to expand this work, hypothesizing that the emerging spectrum of CVM genotypes is associated with different clinical phenotypic manifestations of disease. In Specific Aim 1, cases of inherited arteriovenous malformations are screened for mutations in recently identified HHT genes, and cases of inherited cavernous malformations for CCM genes; the spectrum of clinical manifestations (penetrance, radiological lesion appearance, non-neurologic manifestations, age at clinical presentation and risk of stroke and epilepsy) are compared among cases with different genotypes and with sporadic cases. In Specific Aim 2, lesion histopathology, immunohistochemistry and ultrastructure are systematically compared among various categories of genotype; and in Specific Aim 3, cultured cell lines isolated from the different categories of excised CVM lesions are compared for differential expression of angiogenesis receptors and respective gene products. it is anticipated that this translational research will contribute toward a classification of CVMs based on biologic mechanisms and more rational strategies for screening, predicting, explaining and potentially modifying their clinical behavior.